Andrew Kohut, president of the Pew Research Center, has penned an ignorant op-ed for today’s New York Times about the ostensibly bad polling of the NH Democratic primary. After considering a number of factors that may have affected the poll results, Kohut turns to racism by lower-class whites as an explanation:
To my mind all these factors deserve further study. But another possible explanation cannot be ignored  the longstanding pattern of pre-election polls overstating support for black candidates among white voters, particularly white voters who are poor.
This is the so-called Bradley/Wilder effect. Kohut relies on his own experience polling for former NYC Mayor Dinkins, who won by a two-point margin, arther than the 15% reflected in pre-election polls.
Kohut does not address a fundamental flaw in this thesis — the pre-election polls did not greatly overstate the share of the vote Obama was expected to win. As noted here yesterday:
…the NH polling average for Sen. Barack Obama was 38.3% and the ultimate result was 36% – within the margin of error and thus not supportive of the Bradley/Wilder effect theory.
The Bradley/Wilder theory hypothesizes that certain types of whites will lie to pollsters about their support for a black candidate, but the poll average shows no such exaggeration. The gap is mostly in the underestimation of Clinton. The biggest exception to this may be the Zogby tracking poll, which is addressed below.
Kohut also did not bother to check with the pollsters at issue in this instance, who suggest that the actual results were signalled in the latest polling data.
Rasmussen explained:
In Rasmussen Reports polling, our final trend was in Clinton’s directionâ€â€our tracking poll showed Obama’s lead declining from 10-points following the Sunday interviews to seven points after the Monday night calls. Extrapolating that trend another day would have pointed to a much closer race. Additionally, the Rasmussen Reports surveys showed that Clinton supporters were somewhat more certain that they would stick with their candidate than supporters of Obama or Edwards. If this is the case, why didn’t the late trend get more notice? Perhaps because few other firms polled on Monday night. So, the last polls reported by many continued to show an uptick for Obama.
Zogby also conducted last-minute polling:
My polling showed Clinton doing well on the late Sunday night and all day Monday – she was in a 2-point race in that portion of the polling. But since our methods call for a three-day rolling average, we had to legitimately factor the huge Obama numbers on Friday and Saturday – thus his 12 point average lead. Unfortunately, one day or a day–and–a–half does not make a trend and we ran out of time.
Finally, Kohut’s rush to analyze the income statistics of New Hampshire’s white voters causes him to gloss over a significant point he makes:
Of course these are not the only patterns in Mrs. Clinton’s support in New Hampshire. Women rallied to her (something they did not do in Iowa), while men leaned to Mr. Obama. Mrs. Clinton also got stronger support from older voters, while Mr. Obama pulled in more support among younger voters. But gender and age patterns tend not to be as confounding to pollsters as race, which to my mind was a key reason the polls got New Hampshire so wrong.
Kohut ignores that gender and age were very much factors in this instance. Zogby tells the tale:
Going into the New Hampshire primary, we certainly did see Clinton holding on to a significant lead among women and older voters. But we were focusing on Obama’s massive lead among younger and independent voters. We seem to have missed the huge turnout of older women that apparently put Clinton over the top.
Rasmussen also notes the turnout factor:
Another possibility is that the polls simply understated Clinton’s support. At one level, Clinton’s campaign organization may have been great at getting out the vote. One analyst noted that “The Clinton turnout operation in Manchester their strongest area, was very good, and turnout soared 33% over 2000. In Rochester-Dover-Somersworth, another strong Clinton area, turnout was up 94% from 2000.†That could account for a several percentage points, but not the ten point gap between our final poll and the actual results.
Perhaps, but Rasmussen’s final polling was showing a seven point gap, not a ten point gap. Shaving several points from a seven point gap likely would have put the final numbers within the margin of error. As noted here yesterday, the Clinton campaign had top talent for its ground game in New Hampshire. Kohut does not even consider that the discrepancies in actual turnout for various income groups was driven by the Clinton campaign.
In one sense, this is not surprising. Pollsters (as Kohut was and perhaps still is) make their living asking samples of the public what they think about campaigns. They do not spend nearly as much time asking how candidates go about changing the sample that shows up to vote.
Update: At Just One Minute, Tom Maguire takes on Kohut, noting that Kohut sang a different tune when he did the polling for NH in 1988 and then-Veep George H. W. Bush had a surprise victory over Sen. Bob Dole. I think the kids have a term for it — pwn3d.
Ah, yes, the Pew Research Center – where the motto is, “No Matter What It Is, America’s Always At Fault.”
Chicagoans may remember legendary crusty columnist Mike Royko, who always encouraged his readers to lie to pollsters, making their predictions (and thus election influence) useless. Perhaps some citizens of the Granite State, exhausted after being besieged by repeated election-related telemarketing calls, decided to mess with the Media Election System.
(Actually, that should read “Media/Election Complex.”)
Andrew Kohut is a tool. Can’t link right now but Pew was a big part of the astroturfing that spawned campaign finance reform.
Amazing that folks still think that polls are indicative of some actual result, with high-double-digit probability. Most people I pal around with would tell pollsters to piss off, and there’s a decided slant to our collective voting habits.
He’s completely wrong about this, but I think knowingly so, and shrewdly so, from the perspective of an Obama-pusher.
He’s appealing to Times readers’ (who are pretty much the whole of the Democratic Obama contituency) hatred for their most demonized class, poor men — when you hear “lower-class whites,” you picture men, period; the media has long made sure of this — and concealing the research-supported fact that, basically, women are racist, and vote “identity,” while men aren’t, and men don’t.
That racism does help explain Hillary’s “surprise” win. But it’s sexist to know that.
I’m white. I’d vote for a black candidate. There aren’t any running for President though. Unless Hillary is considered “the first black ex-first lady”.
Hillary is running for Lioness. And all who need the “wonderful opportunities” that she has for them, will be her cubs. She is their mother and she aches for them. She aches so much that she would walk through fire to protect them. It makes her cry sometimes.
Would a lioness, whose cubs are threatened, resist busing 8000 voters over from MA?
The Simba Theory
Polls, shmolls.
I don’t know how many people are like me, but on the rare occasion that I am polled, I give all opposite answers. You can even tell the pollster that you have no idea what is going on, and they poll you ANYWAY!
I just resent the hell out of “polls” becoming election drivers, and the drive-bys using whatever small portion of the polls that fit their purpose.
Has anyone ever heard a major media outlet even mention “internals”?
My guess is that about 80% of Americans wouldn’t even know what you were talking about if confronted with the word “internals”.
Can’t we just let the man call the Democratic base a bunch of racists without rebutting him? Sure, it’s a crock of steaming shit, but it’s fun to watch them eat their own. I, for one, am enjoying my little vacation from Teh H8ing while this goes on.
Well, I mean they can’t admit that polling is crap and their efforts were worthless. It has to be someone else.
Excellent point. I do love a dustup between two grievance groups. If we could get the environmentalists involved, I’d be giggling like a Japanese schoolgirl.
How soon they forget – George Bush beat Dole by 9 points in 1988 after the Monday Gallup Poll had Dole up by 8. Geez, a 17 point swing! And the Head of Gallup at the time, left to explain this to the press – Andrew Kohut.
I’m a middle-aged white male, and I voted for Alan Keyes once. Does this count as having voted for a Black man?
I remeber the polls saying (and me believing) that the Carter-Reagan election was a toss-up. So they’ve had problems at least as far back as 1980.
bergerbilder,
As to your question, the answer is yes, unless you ask someone on the Left.
As to the 1980 election, the poll problem there was with John Anderson, as people incorrectly believed he would draw votes from Reagan. Reagan won, but with 50.7% of the popular vote. In that sense, the election was pretty close, though 2000 demonstrated it can be closer.
Nope. You might as well have casted a vote for Simple Sambo.
KAAAAAARRRRRRLLLLL!
Karl,
I completely forgot about Anderson, thanks for reminding me.
It’s not likely that I’ll forget Ross Perot as quickly, though.
Slart,
That’s one of my favorites. If you haven’t seen Free Enterprise, you really should.
And if you have, the fact that the director now does a lot of the “behind the scenes” extras for DVDs will be twice as amusing.